Representing Discourse Coherence: A Corpus-Based Study
By Florian Wolf, Edward Gibson. International Conference on Computational Logic, 2005
307 citations recorded by the source corpus.
This article aims to present a set of discourse structure relations that are easy to code and to develop criteria for an appropriate data structure for representing these relations. Discourse structure here refers to informational relations that hold between sentences in a discourse. The set of discourse relations introduced here is based on Hobbs (1985). We present a method for annotating discourse coherence structures that we used to manually annotate a database of 135 texts from the Wall Street Journal and the AP Newswire. Alltexts were independently annotated by two annotators. Kappa values of greater than 0.8 indicated good interannotator agreement. We furthermore present evidence that trees are not a descriptively adequate data structure for representing discourse structure: In coherence structures of naturally occurring texts, we found many different kinds of crossed dependencies, as well as many nodes with multiple parents. The claims are supported by statistical results from our hand-annotated database of 135 texts.
Subfields: Artificial Intelligence
View Paper: https://www.semanticscholar.org/paper/16fee45352cad81fae2dd58b3bd9624eb1141c86
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
Citations in the corpus
- A Hierarchical Neural Autoencoder for Paragraphs and Documents (2015, background): “18 can be interpreted as follows: ( j − i ) denotes the distance in terms of position index be-tween two outputted sentences indexed by j and i , 3 Wolf and Gibson (2005) and Lin et al. (2011) proposed metrics based on discourse relations, but these are hard to apply widely since identifying…”
- Discourse-Aware Neural Extractive Text Summarization (2019, methodology): “Other alternatives includes Graph Bank (Wolf and Gibson, 2005) and PDTB (Miltsakaki et al., 2004).”
- Automatic sense prediction for implicit discourse relations in text (2009, methodology): “However, GraphBank annotations do not differentiate between implicits and explicits, so it is difficult to verify success for implicit relations.”
- The Penn Discourse TreeBank 2.0. (2008, background): “, 2001) and GraphBank (Wolf and Gibson, 2005) is that the resulting structures are unecessarily complex.”
- DialogSum: A Real-Life Scenario Dialogue Summarization Dataset (2021, background): “First, a big challenge posed by spoken dialogues is that their information flow is different from monologic text, which is intuitively reflected in the dialogue discourse structures (Wolf and Gibson, 2005).”
- TIARA: A Tool for Annotating Discourse Relations and Sentence Reordering (2020, background): “A text is usually represented as a tree (Mann and Thompson, 1988; Stab and Gurevych, 2014) or graph (Wolf and Gibson, 2005; Sonntag and Stede, 2014; Kirschner et al., 2015).”
- Machine Comprehension with Discourse Relations (2015, background): “Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, pages 1253–1262, Beijing, China, July 26-31, 2015. c©2015 Association for Computational Linguistics”
- Revisiting Readability: A Unified Framework for Predicting Text Quality (2008, background): “In the past, subsections of the Penn Treebank (Marcus et al., 1994) have been annotated for discourse relations (Carlson et al., 2001; Wolf and Gibson, 2005).”
- HiEve: A Corpus for Extracting Event Hierarchies from News Stories (2014, background): “Event mentions in news stories are, arguably, particularly spatiotemporally vague, because news stories put more focus on structure and coherence of the narrative (Wolf and Gibson, 2005) than on spatiotemporal properties of the individual events.”